We are developing techniques for cutting serial sections of plastic embedded biological material that are thinner than have been previously cut. We have purchased a modified Ultracut-UCT microtome from Leica, Inc., and are working with Diatome U.S. to develop experimental diamond knives to investigate the practical limitations to section thickness. These tools, in conjunction with an isolation room to minimize air currents, a vibration damping table and proper orientation and trimming of the sample block face, have allowed us to obtain serial sections that are less than 11nm in thickness. These sections cut with reasonable uniformity (q10%) and may be obtained in large numbers (>100). By staining sections with methanolic uranyl acetate and using a small objective aperture in a conventional TEM operating at 40 or 60kV, we are able to obtain good image contrast. Experimentation with diamond knives that have decreased included angles (<40x) has resulted in noticeable improvements in section quality, with respect to chatter and compression. The microtome and diamond knives no longer appear to be the limiting factors in section thickness, so we are now turning our attention to other factors, such as sample preparation, embedment hardness and block face design. Through this work we hope to learn how thin a set of serial sections can be cut.